What we checked on Kore.ai
Multi-agent
The runtime ships four named orchestration patterns, so several specialist agents can hand off, delegate, fan out or talk across runtimes on one task.
Compose specialists with HANDOFF, DELEGATE, FAN_OUT, and A2A — typed payloads, governed handoffs, unified traces.
docs.kore.ai/homeChecked
Agent permissions
Policies and constraints are enforced by the engine rather than by the model, and guardrails can block a tool call, a tool result or a handoff before it happens.
Policies are enforced outside the LLM, so agents can reason while staying within approved business, security, and compliance boundaries.
kore.ai/ai-agent-platformChecked
Open models
Open-weight families are in the supported-model catalogue through Groq, Fireworks, Together and Microsoft Foundry, including Llama, Mixtral, Qwen and DeepSeek distills.
OpenAI, Anthropic, NVIDIA, Cohere, LLaMA 2, and additional providers. Open-source and custom fine-tuned models are supported through the same…
Model choice
You pick the models: agents reference Fast, Balanced, Powerful or Voice tiers and admins bind each tier to a model from any registered provider.
The Agent Platform is provider-neutral. Agents can use models from multiple LLM providers, and you can switch between them without changing agent…
Self-hosted
Besides the SaaS default, the vendor documents on-premises and private VPC deployment, offered to enterprise customers with residency or compliance needs.
Self-hosted (Enterprise) | For organizations requiring on-premises deployment. Same platform capabilities, available for enterprise customers with…
SDK languages
The documented SDK set is client-side only: a JavaScript/TypeScript Web SDK plus React Native, Flutter, Android and iOS mobile SDKs, with no server-side SDK listed.
The Web SDK is a JavaScript/TypeScript SDK that provides programmatic access to chat and voice interactions, including streaming responses, message…
Tracing and evals
Every session is traced in structured detail and evaluation suites run before and after release, with the vendor claiming full rather than sampled coverage.
Captures over 200 trace events per session. Includes runtime inspection, alerting, and session-level debugging across multi-agent interactions.
How it meters
Billing runs on a subscription tier plus workspace credits and billing units you can top up, with your own model spend metered separately per provider key.
Credits allocated to your workspace.
Free tier
There is no self-serve way in: sign-up is gated to approved domains and the quick start tells everyone else to contact the company for access.
Agent Platform is available to Kore.ai customers and partners. Sign-up is enabled only for approved domains.
MCP server
Kore.ai ships its own stdio MCP server, so an outside assistant such as Claude Code or Cursor can build, evaluate and analyse agents on the platform.
Agent Platform exposes its features through a Model Context Protocol (MCP) server. Your AI assistant can connect to the platform and perform platform…
Bring your own key
You register your own provider API keys for inference, and may also supply your own encryption keys from AWS, Azure, GCP or an external KMS on the Enterprise plan.
LLM provider credentials: API keys for LLM providers (for example: Anthropic, OpenAI, Azure, Google, Bedrock).
A company can buy it
This is bought by organisations, not individuals: an optional organization layer carries shared billing and governance over several workspaces.
Organization | Optional. It sits above workspaces and carries shared billing and governance across several teams. A single-workspace setup doesn't…
Seat model
Access is granted per invited member with a workspace role, but no seat price, seat floor or seat ceiling is published anywhere I read.
Invitations carry the role you choose, so decide the role before you send one rather than correcting it afterward.
Pooled budget
Credits are held as one workspace balance that projects draw on, with token budgets and rate limits set at workspace level rather than per user.
Which AI models and providers your team can reach, through LLM policies, token budgets, and rate limits.
Admin controls
Admins constrain agents centrally: project-level guardrail policies apply to every agent, and a workspace approval policy gates production deployments.
Project guardrails | Centralized governance and reusable safety policies across agents. | Govern > Guardrails | Enterprise-wide safety enforcement.
Audit log
Administrative action is recorded per tenant with actor, target, IP and trace, filterable by category; CSV export and 90-day retention need Professional or Enterprise.
Audit logs give you a tenant-scoped, filterable record of every significant workspace action. Use them to investigate incidents, confirm compliance,…
Single sign-on
Both SAML 2.0 and OpenID Connect are supported, with JIT provisioning and group-to-role mapping, but the whole SSO tab is gated to the Enterprise plan.
SSO lets your users authenticate through an external identity provider. Plan requirement: Enterprise plan only.
Opt out of training
No opt-out is needed: the privacy notice states customer data is used only to deliver the service and never to train generalised models or improve the product.
Kore.ai does not use that Customer Data for internal training, knowledge-base updates, internal product enhancements or updates, or to train…
kore.ai/privacy-policyChecked
Data residency
Thirteen regional clouds are named, from the UK and Germany to Japan, Saudi Arabia and India, and on-premises or private VPC deployment is offered as well.
The platform runs in multiple regions.
Getting out
Agent definitions are Git-native YAML artifacts you hold in version control, and the audit log exports to CSV; credit expiry is nowhere stated.
Agent definitions are immutable YAML artifacts that are Git-native, diffable, and reviewable in pull requests.
Certifications
The enterprise foundation is stated to carry PCI, SOC 2, GDPR and FedRAMP certifications, with KMS compliance profiles selectable per tenant.
Provides versioning, environment management, and certifications for PCI, SOC 2, GDPR, and FedRAMP.
Kore.ai price and terms
Terms
- Free plan or trial
- NoNo
What needs a higher plan
- Enterprise
- Single sign-on
For AI agents
- MCP server
- Yes
Where these prices come from
koreai.mintlify.site/agent-platform/administration/workspace-account.md, checked
Kore.ai alternatives
All 9 alternatives- This spot is openBe the first agent people see on Kore.ai's page.
- First on your rivals' pages
- Top of your category
- First in the home page list
- First in search results
- In comparisons and guides
- In the weekly newsletter
- Visits and AI reads each month
7 more alternatives
8 of the 9 publish a price.
- GumloopFrom $37 a monthMultiplayer AI agent builder: teams build agents over their apps while IT keeps governance and spend control.
- DustFree, then about $34 a seat a monthPlatform for building and running company agents on connected tools, data and 20+ models.
- DifyFree, then $59 a monthOpen-source platform for building agentic workflows, AI agents and RAG knowledge pipelines.
- ManusFree, then $20 a monthGeneral-purpose AI agent that plans and executes multi-step tasks on its own cloud computer.
Questions about Kore.ai
- How much does Kore.ai cost?
- Kore.ai gives its price on request. From koreai.mintlify.site/agent-platform/administration/workspace-account.md, checked 28 October 2026.
- Does Kore.ai have an API or an MCP server?
- It has an MCP server.
- Is Kore.ai open source, and can you host it yourself?
- You can host it yourself.
- What have we checked on Kore.ai?
- 21 facts, under these headings: How much it does unattended, which models it runs on, whose machine it runs on, building with it, what it costs to run, buying it for a team, what happens to your code.
- How do we check Kore.ai?
- We hold 21 facts on Kore.ai, quoted from 11 pages. Facts and price last checked 28 October 2026.
Sources and changes
Last changed
Sources11
- koreai.mintlify.site/agent-platform/administration/workspace-and-team.mdIndependent source
- koreai.mintlify.site/agent-platform/index.mdIndependent source
- koreai.mintlify.site/agent-platform/administration/workspace-account.mdIndependent source
- koreai.mintlify.site/agent-platform/getting-started.mdIndependent source
- koreai.mintlify.site/agent-platform/administration/ai-configuration.mdIndependent source
- koreai.mintlify.site/agent-platform/mcp-server.mdIndependent source
- koreai.mintlify.site/agent-platform/safety-and-guardrails.mdIndependent source
- koreai.mintlify.site/agent-platform/sdk/overview.mdIndependent source
- kore.ai/ai-agent-platform
- docs.kore.ai/home
- kore.ai/privacy-policy
Is this yours?
AI assistants like ChatGPT and Claude opened The Agents Index pages 2,746 times in September. Claiming is free. You get this page's reader numbers and the verified owner mark. You confirm it with an email at kore.ai.
Report an error if a fact is wrong.